Modeling Parental Provisioning by Red-winged Blackbirds in North Dakota

University of Nebraska - LincolnDigitalCommons@University of Nebraska - LincolnUSDA National Wildlife Research Center - StaffPublications Wildlife Damage Management, Internet Center for

1-1-2011

Modeling Parental Provisioning by Red-wingedBlackbirds in North DakotaGeorge M. LinzUSDA–APHIS–WS, george.m.linz@aphis.usda.gov

Richard S. SawinDepartment of Biological Sciences, North Dakota State University, Fargo, ND 58105, USA

Mark W. LutmanUnited States Department of Agriculture, Mark.W.Lutman@aphis.usda.gov

William J. BleierNorth Dakota State University, William.Bleier@ndsu.nodak

Follow this and additional works at: http://digitalcommons.unl.edu/icwdm_usdanwrc

This Article is brought to you for free and open access by the Wildlife Damage Management, Internet Center for at DigitalCommons@University ofNebraska - Lincoln. It has been accepted for inclusion in USDA National Wildlife Research Center - Staff Publications by an authorized administratorof DigitalCommons@University of Nebraska - Lincoln.

Linz, George M.; Sawin, Richard S.; Lutman, Mark W.; and Bleier, William J., "Modeling Parental Provisioning by Red-wingedBlackbirds in North Dakota" (2011). USDA National Wildlife Research Center - Staff Publications. Paper 1335.http://digitalcommons.unl.edu/icwdm_usdanwrc/1335

The Prairie Naturalist 43(3 I 4):92-99 ; 2O11

Modeling Parental Provisioning by Red-winged Blackbirds in North Dakota

GEORGE M. LINZI, RICHARD S. SAWIN', MARK W. LUTMAN3, AND WILLIAM J. BLEIER

United States Department of Agriculture, Wildlife Services, National Wildlife Research Center, 2110 Miriam Circle,Suite B, Bismarck, ND, USA 58501 (GML)

Department of Biological Sciences, North Dakota State University, Stevens Hall, Fargo,' ND, USA 58105 (RSS, MWL, WJB)

ABSTRACT Maie red-winged blackbird s (Agelaius phoeniceus)exhibit a difference in nest provisioning rates along an east-westgradient,in North America. North Dakota is located in the center of North America and harbors a large population of breedingred-winged blackbirds (RWBL). This location provided an opportunity to compare provisioning rates in the central U.S. withthose reported for the easterrr and western populations. We placed video cameras at RWBL nests to record male and femalefeeding trips. Thirty-four nests were located on territories with original males and 30 were on ter:ritories where a replacement(floater) male had taken over a territory after we removed the original male by shooting or trapping. Original territory ownerswere more likely to feed young, and males were more likely to feed at nests with older chicks, at nests with more chicks, and laterin the breeding season. Red-winged blackbird parental provisioning patterns in North Dakota appear to be similar to thoseobserved in eastern North America.

KEY,WORD S Agelaius phoeniceus,nestlings, North Dakota, parental care, provisioning, red-winged blackbird

The amount and quality of parental investment isimportant because it can contribute directly to the survivalof offspring (Westneat and Sherman 1993). Most songbirdsare monogamous, and both males and females contributelarge amounts of parental care (Lack 1968). For a fewspecies, however, breeding in highly productive habitats canallow one parent to gather sufficient resources to feed(Beletsky 1996). This condition, in concert withanisogamy, means that the benefits of male provisioningmight be reduced. For these species, the amount of maleprovisiqning may be variable. Polygyny is expected to arisewhen males can decrease parental provisioning with littlecost and increase their reproductive success by attractingadditional mates (Webster 1991).

Interest in male red-winged blackbird (RWBL; Agelaiusphoeniceus) provisioning has been augmented by noticeabledifferences in provisioning rates along an east-west gradientthat may be due to differences in habitat productivity, haremsize, or breeding season length (Muldal et al. 1986, Beletsky1996). For example, in Washington, only 67o of males fednestlings, whereas 29-88Vo of males in the eastern UnitedStates feed nestlings (Searcy and Yasukawa 1995).However, Whittingham and Robertson (1994) found that337o of the males in an Alberta population fed nestlings, andpresented additional evidence that the east-west gradient inmale provisioning rates may not be as strong as previouslythought.

North Dakota lies between the eastern and westernRWBL populations and thus, provides an opportunity tocompare provisioning rates among these populationi.Specifically, our objectives were to (1) document parentalprovisioning by RWBL for a population in North Dakota,(2) experimentally determine the role of breeding

experience on male provisioning, ald (3) examine theresponse of females to male assistance.

STUDY AREA

From 29 May to 29 lune 2000-2001, we observedfeeding contributions by male and female RWBL at 10

wetlands in Barnes County, North Dakota (46.93" N, 98.24oW), which was in the central portion of the Drift PlainsEcoregion (Bryce et al. 1998). This agricultural area

contained a mix of row crops and Conservation ReserveProgram (CRP) grassland interspersed with numeroustemporary and seasonal wetlands. The Drift PlainsEcoregion hosted a large population of RWBL that breed inboth wetland and upland habitats (Nelms et al. 1999). Our10 study marshes contained a variety of wetland habitatsincluding those covered by emergent vegetation, open waterringed by cattall(Typha sp.), and roadside ditches.

METHODS

To study the effects of breeding experience on maleprovisioning, we manipulated territory ownership on each ofthe study wetlands. As a polygynous species, male RWBLestablished exclusive territories and attempted to attractmultiple females. Because the overall sex iatio for RWBLapproached 50:50, this arrangement left a pool of floatermales who regularly monitored territory ownership, lookingfor a chance to establish a territory in a vacant area. Onedifficulty with territory manipulations involving RWBL isthat neighboring territory owners often annex the territoriesof removed males, preventing floaters from filling vacancies(Beletsky and Odans 1996). This complicated any effoft to

1 Corresponding author email address: George.m.linz@ aphis.usda.gov.2Present address: 3405BBaywood Lane, Greenville, NC 27834,U5A.3 Present address: USDA, National Wildlife Research Center, 4101 LaPorte Avenue, Fort Collins, CO 80521, USA.

The Prairie Naturalist' 43(314): December 2011

manipulate ownership of individual territories.To circumvent this problem, while still controlling for

variation between wetlands, we used a box trap with a livedecoy bird and shooting (permit #M8019065-5) topermanently remove territory owners on a randomlyselected half of each study wetland (Bray et al. 1975). Priorto removal, we captured all tenitorial males and fitted themwith a United States Fish and Wildlife Service aluminumband (permit # 21672) and a unique combination of plasticcolor bands in order to track movements into vacatedterritories. The original territory owners remained in placeon the other half of the wetland. We also captured andbanded the replacement males on the manipulated portion ofthe wetland. The result of the manipulations was that halfof the territory owners resembled the overall population(control), with a mix of experienced and new breeders. Theother half (treatment) contained floaters that moved into thevacated territories. Most of these floater males probablynever maintained a territory and had no breeding experience(Beletsky and Orians 1996). Removals occurred beforemost females had started to build nests. Throughout theexperiment, we tracked active nests by searchiirg'for.newnests every 6 days and checking on the status of knownnests every 3 days.

Accurately observing parental provisioning in person can

be difficult because real-time observations are timeconsuming, observer presence can alter feeding patterns,and feeding can be hard to observe at obscured hests.

Therefore, we used small video cameras mounted on tripodsand concealed with camouflage netting to record feedingactivities at randomly selected nests containing nestlings.We were able to aVoid recording at two nests on the sameterritory because most males were color-banded, and wemade regular territory maps. We typically recorded at twonests each day. When possible, we selected one nest from a

treatment area and one nest from a control area in order tocollect data in treatment and cbntrol areas throughout theseason.. Recording did not take place when weatherconditions would interfere with camcorder electronics.

To allow the blackbirds to adjust to the presence of thecamera, we placed camouflage netting 1-3 m from the nest12-24 fu before recording began. For each day ofrecording, we selected a Z-hr time block (the length of a

videotape) at random from the first and second half of theday (AM and PM), with the stipulation that the tworecording periods were at least 2.5 hr apart. At th.e

designated recording time, we placed the video cameras

under the pre-positioned camouflage netting and orientedtowards the nest. The investigators left the area until afterthe 2-br recording period was complete.

After videotaping was complete, we viewed tapes in a

random order, and the observer did not know the location oridentity of the nest. For each feeding visit, we recordedtime of visit and sex of the provisioning parent. Oneadvantage of videotapes over real-time observations is thatnest visits without feedings can be verified. For purposes of

93

analysis, we considered AM and PM recordings from thesame day samples of the same experimental unit and

averaged. In order to minimize the effects of camera setup,we did not use the first 15 min of each tape. In practice, thefirst female feeding typically occurred before the 15 minperiod had passed.

We considered male feeding as a binary response(yes/no), so some of the possible factors influencing theprobability of male feeding were examined using logisticregression (Agresti 1996). We considered 6 a priori modelsranging in complexity from a similar probability for allmales to a dependence on treatmenVcontrol status, numberof chicks, age of chicks, and Julian date (Table 1). Thetreatment/control factor loosely represented breedingexperience; males in treatment areas were expected to.havevirtually no breeding experience, while control areas

resembled the overall population containing a mix ofexperienced and inexperienced breeders.

The simplest model to describe male feeding probabilityassumed a constant feeding probability for all males. Asecond model, (T), assumed a difference based primarily ontreatment/control status. Based on consistent reports thatmales only feed chicks older than 4 days, all 4 of theremaining models included nestling age as a factor. Supportin previous studies for Julian date and number of chicksvaried, so a combination of models including these factorswero considered (Table 1).

We used odds ratios for determining associationsbetween chick feeding (response) and predictor variables,which included treatment (original territory owners orreplacement territory owners), nestling age, number ofchicks and date. The odds ratio for a predictor variable was

the relative amount by which the odds of the outcomeincrease (odds ratio >1.0) or decrease (odds ratio <1.0;Hosmer and Lemeshow 2000, Jacques et al. 2011). That is,multiplicative effects on the odds of a l-unit increase in the

response variable were associated with fixed leveld of otherpredictor variables (Hosmer and Lemeshow 2000, Jacques

et al. 2011).We summarized the data for female feeding for each nest

as the number of feeding trips/hour and used log-linearmodels (log link/ Poisson distribution) to relate feeding tripsto male assistance (yes/no), number of chicks, and age ofchicks (Agresti 1996). We considered 7 a priori models,differing by which potentially influential factors wereincluded (Table 2). The first model, H, modeled the numberof female feeding trips/hour as a log-linear function of maleassistance. The remaining models considered effects of the

number and age of chicks while including or excluding maleassistance. Support for models that included maleassistance (Table 2; models H, HA, HC and HAC) couldindicate that females scaled back their feeding efforts whenmales were actively feeding. Conversely, males may haveinitiated feeding when female provisioning was notsufficient. Support for models excluding male assistance(Table 2; models A, C and AC) may have indicated that

94

females maintained feeding effort regardless of the level ofmale assistance. This implied that male feeding was inaddition to female feeding and resulted in an increasedoverall level of provisioning.

We considered each of the candidate models to representa competing hlpothesis regarding the factors influencingparental provisioning in this population. Our overall goalwas to estimate feeding-related parameters (Stoehr i999);using a model with. too few parameters would introducestatistical bias, and using too many parameters wouldunnecessarily increase the variance of our estimates.Therefore, we used a corrected Akaike's InformationCriterion corrected for small sample size (AIC.) to select aparsimonious model (or set of models) to make inferencesabout parental provisioning (Burnharn and Anderson 2002).We performed statistical analyses using the LOGISTIC(males) and GENMOD (females) procedures in SAS v.8.0(SAS Institute, Inc. 1999).

RESULTS

_Across study years, we videotaped parental feeding for256 hr at the nests of 64 different males. We did notidentify individual females; however, the number ofpotential females and the time required to re-nest andincubate make it unlikely that any females were observedmore than at a single nest. Nests varied in the number andage ofchicks present (Table 3).

We recorded male provisioning at 15 nests located at 8 ofthe 10 study wetlands. Males provisioned nests with as fewas 1 chick (26 June 2000,3-5 days old) and with chicks asyoung as 0-2 days old (5 chicks ,29 lane 2000). Averagefemale provisioning rates (across all combinations of ageand number of chicks) were similar in treatment (957o CI =89-12.6 trips/hr) and control arcas (95Vo CI = 8.4-11.5trips/hr). Males in the control area made far fewer feeding

Linz et al.' Parental Provisioning in Red-winged Blackbirds

trips than did females (95Vo CI = 1.8-3.9 trips/hr).Four of. the 6 models considered to describe the

probability of male feeding had substantial support (Table1). Model TA had 33Vo of the support, but none of themodels had significant support over the others. An analysisof the odds ratios for each of the factors shows that'treatment group' has the most dramatic effect; theestimated_odds of an original territory owner feeding chicksare about 20 times the odds of a replacement owner feedingchicks (Table 4). Following with the concept of taperingeffect sizes, the age of chicks may be. the next importantfactor. Males were estimated to be about 2.8 times morelikely to feed for each 3 days of growth. These modelsdescribed smaller impacts for number of chicks and Juliandate. For Julian date, however, the odds of male feedingincreased 57o for each day that passed, a substantial increaseover the coirrse ofthe breeding season.

We would expect male feeding rates in the overallRWBL population in this region to be similar to the ratesobserved in the control side of our study. Rates for aparticular circumstance can be determined based on theparameters given in Table 5. For example, under model TAthese data would predict that 16.5Vo of the males would feed0-2 day-old chicks and 59.8Vo would feed 6-8 day-oldchicks (Fig. 1). We think these equations are more usefulfor comparing between populations than citing only theproportion of males in our population that fed nestlings.

Only two of the models considered to describe femalefeeding had substantial support (Table 2). Borh modelsincluded number of chicks and age of chicks, confirming theexpected influence of these factors on female feeding rates.Model AC gained almost twice as much support as modelHAC (65Vo vs. 35Vo). The estimated number of feedingtrips for females in any situation can be reconstructed withthe parameters shown in Table 6.

Table 1. Summary of the evidence supporting each of 6 a priori logistic regression models considered to describe the probabilityof observing a male feeding at red-winged blackbird nests in Barnes County, North Dakota, USA, 2000-2001.

TA

TAD

TACD

T

Constant

46.93

45.86

44.36

54.84

69;10

52.55

53.24

53.42

58.65

11.63

0.00

0.69

0.87

6.10

19.08

0.33

0.24

a.2r

0.02

0.00

nT=Treatment(T)onControl(C),A=Nestlingage,C=Nu*b"iInformation Criterion for small sample size (Burnham and Anderson 2002); o Difference in AIC. relativi to minimum NC.i "Akaike weight (Burnham and Anderson 2002).

The Prairie Naturalist ' 43(314): December 2011

Table 2. Summary of the evidence supporting each of 7 a priori log-linear models considered to describe the number of femalefeeding trips/hr at red-winged blackbird nests in Barnes County, North Dakota, USA, 2000-2001.

95

Modelu Loglikelihood If AIC"" AAIC". eWi

AC

HAC

H

A

C

HA

HC

902.80

903.05

884.73

888.19

896.58

888.22

896.84

-1799.96

-1798.72

-t765.65

-1772.57

-t789.35

-t770.81

-1788.05

0.00

1.24

34.3r

27.39

10.61

29.15

11.91

0.65

0.35

0.00

0.00

0.00

0.00

0.00

uT =Treatrirent (T) orControl (C),A-Nestling og€, C = Numberof nestlings,D =Date;

oNumberof parameters; "Akaike'sInformation Criterion for small sample size (Burnham and Anderson 2002);d Difference in AIC" relative to minimum NC"i "

>-)rl).--.-f-t-cgA-olia!.-o0-H.-Ec)eq)

-c!=a

Akaike weight (Burnham and Anderson 2002).

0.75

0.25

Nestling age

Figure 1 . Functional form of logit linear model TA describing ma,le red-winged blackbird (Agelaius phoeniceus) feedingprobability for original and replacement territory owners as a function of nestling age in Barnes County, North Dakota, USA,2000-2001.

0.5

Original Territory Owners

96

DISCUSSION

Male RWBL in central North Dakota fed nestlings on aregular basis. While the data did not clearly arbitratebetween some of the competing models, the importance ofmale breeding experience was clear. The importance ofnestling age, number of chicks, and Julian date, factorspredicted based on theory and previous studies, wereprobably real but less important than male breedingexperience. The proportion of males in central NorthDakota (98' V/ longitude) that fed nestlings was moresimilar to previously studied populations in eastern NorthAmerica (74'W - 89'W) than populations in westerrr NorthAmerica (1i3'W - 123" W; Whittingham aad Robertson1994, Searcy and Yasukawa L995).

Table 3. Chaiacteristics of 64 red-winged blackbird nestsCounty, North Dakota, USA, 2000-2001.

Linz el al. . Parental Provisioning in Red-winged Blackbirds

Female provisioning rates (feeding tnps/h) were similarto those reported from other areas (Muldal et al. 1986,Teather 1992, Clark and Lee 1998). However,interpretation of these rates is difficult because the amountof food brought per trip may vary by sex and habitat(Whittingham and Robertson 1994). Variarion in theamount of food delivered per trip adds variation to the dataand could make it more difficult to estimate the effects ofthe factors of interest. Further, the number of feeding tripsper hour does not reflect the changes we would expect basedon the age and number of chicks in the nest. Buildingstatistical models to describe how the number of feedingtrips varies according to additional factors could helpdetermine what influences the number of feeding trips, andwill help facilitate comparisons between populations.

videotaped to study parental provisioning of nestlings in Barnes

Male Feeding

Treatment Group

Original Territory Owners

Replacement Jerritory Owners

Nests Videotaped Males Observed Feeding

l434

30

Chicks/Nest"

Treatment Group

Original Territory Ow.ners

Replacement Territory Owners

L J

l1

5

l1 11

Age of Chicks (days)

Treatment Group

Original Territory Owners

Replacement Territory Owners

0-2 3-5 6-8 9+

1110

t6

Nestlings/nest included red-winged blackbird and brown-headed cowbird chicks.

Our experiment to manipulate male breeding experiencedemonstrated that original territory owners are more likelyto feed nestlings than males that replaced original dwnersfollowing permanent removals. We observed only one malefeeding nestlings in a treated (manipulated) area of thewetlands, where we assumed that males had no breedingexperience. The estimated proportion of feeding males incontrol areas, however, indicated that some males withoutbreeding experience may feed nestlings. Because 40-60Vo

of adult RWBL die each year (Yasukawa and Searcy 1995),we would expect the maximum proportion of males thatfeed nestlings to be in the 40-60Vo range if first yearbreeders do.not feed. Our data predicted situations wherehigher proportions of males may feed nestlings as have beenobserved in other populations (Searcy and Yasukawa 1995),suggesting that males without breeding experience may. feedin some populations.

The P.rairie Naturalist ' 43(3/4): December 2011

Table 4. Estimated odds ratios (and confidence intervals) for factors affecting the probability of observing a male feeding at red-winged blackbird nests for 4 competing models in Barnes County, North Dakota, USA, 2000-2001.

ModeluFactor

Treatment/Control Age of Chicks Number of Chicks Julian Date

97

TA

TAC

TAD

TAC D

2t.t5 2.74

(2.3t,t93.96) (r.26,5.95)N/A

t.3320.74 2.88

(2.26 ,tg0.26) (t.25 ,6.41) (0.70 ,2.s4)

20.r3 2.79N/A

1.52

(0.16,3.02)

(2.r4 ,188.9s) (1.26 ,6.18)

2.99

(2.04 ,172.0t) (r.29 ,6.97)

t8.74

N/A

N/A

. 1.05

(0.e6 , 1.1s)

r.o7

(0.97 , 1.18)

T = Treatment (T) or Control (C), A = Nestling age, C = Number of chicks, D = Date.

Table 5. Estimated parameters (and standard errors) for 4 logistic regression models describing the probability of obserying a

male feeding at red-winged blackbird nests in Barnes County, Nofih Dakota, USA,2000-2001.

Modelu Parameter

lntercept Treatment / Control Age of Chicks Number of Chicks Date

TA

TAC

TAD

TACD

-5.70

(1.s3)

-6.75

(1.e8)

-13.87

(8.36)

-r7.92

(e.0s)

3.05

(1.1 3)

3.03

(1.13)

3.00

(1.14)

2.93

(1.13)

1.01

(0.3e)

1.06

(0.41)

1.03

(0.41)

1.10

(0.43)

N/A

0.29

(0.33)

N/A

0.42

N/A

N/A

0.05

(0.0s)

0.06

(0.3s) (0.0s)

u T = Treatment (T) or Control (C), A = Nestling age, C = Number of chicks, D = Date.

Our results supported the hypothesis that females evidence, however, to rule out the possibility that male

receiving male assistance provisioned at the same rate as RWBL feed when female provisioning was insufficientfemales without male help.. There was not sufficient (Whittingham 1989).

rl

Table 6' Estimated parameters (and standard errors) for 2 log-line* -:g:1, describing the number of female feeding trips/hrred-winged blackbird nests in Barnes counry, Norrh Dakora, tisa, znio_zoot.

Modelu Parameter

Male Assistance Age of ChicksIntercept

Number of Chicks

AC

HAC

r.51

(0.1s)

1.50

(0.1s)

N/A

4.01

(0.10)

0.14

(0.04)

0.15

(0.04)

0.18

(0.03)

0.i8

(0.03)uT = Treatme

ACKNOWLEDGMENTS

M. Biondini, L. Flake, C. Lehman, and G. Nuechterleinprovided suggestions on earlier drafts of this manuscript. L.Penry formatted the manuscript. This projeci wasconducted with the approval of the United States

!g911mint of Agriculture, Wildlife Service, NationalWildlife Research Center,s and North Dakota StateUniversity's (NDSU) Institurional Animal Care and UseCommittee and under conditions outlined by a scientificcollecting permit (#MB 019065). Funding was providedjointly by the National Wildlife Research i"nt", (protocolQA-803), a unit within the Wildlife Services program of they"ir"d States Department of Agriculture, Animal and plant

I:dth Inspection Service, and by the Deparrmenr ofBiological Sciences ar NDSU.

LITERATURE CITED

Agresti, A. 1996. An introduction to categorical data'' analysis. John Wiley and Sonr, N"*-Voif., N"*York, USA.

Beletsky, L. D. 1996. The red_winged blackbird: rhebiology of a strongly polygynous songbird.Academic press, San Diego, California, USA.

Beletsky, L. D., and G. H. Orians. 1996. Red_wingedblackbirds: decision making and reproductivesuccess., University of Chicago press, Chicago,Illinois, USA. ,

Bray, O. E., J. L. Guarino, and W. C. Royall, Jr. l9jl. Ajqap for capturing territorial male red_wingedblackbirds. Western Bird Bander S0:4_:l .

Bryce, S., J. M. Omernik, D. E. pater, M. Ulmer, J. Schaar,J. Freeouf, R. Johnson, p. Kuck, and S. H.Azevedo. 1998. Ecoregions of Nofth Dakota andSouth Dakota. Northern prairie Wildlife ResearchCenter Home page. http://www.npwrc.usgs.gov/

resource/1999/ndsdeco/ndsdeco.htm (Version 3CNov.98).

Burnham, K. p., and D. R. Anderson. 2002. Modelselection and multimodel inference: a practicalinformation-theoretic approach. Second Edition.Springer-Verlag, New york, New york, USA.

Clark, A. 8., and W. Lee. 1999. Red-winged blackbirdfemales fail to increase feeding in response tobegging call playbacks. Animal Behaviour56:563-570.

.Hosmer, D, W., and S. Lemeshow. 2000. Applied logisticregression. Second edition. John Wiley and-SonsIncorporated publication, New york, New york,USA.

Jacques, C. N., T. R. Van Deelen, W. H. Hall, Jr., K. J.Martin, and K. C. Vercauteren. 2071. Evaluatinghow hunters see and react to telemetry collars on. white-tailed deer. Journal of Wildlife Managementl5:221-231.

Lack, D. 1968. Ecological adaptations for breeding inbirds. Methuen, London, England.

Muldal, A. M., D. Moffatr, and R.l. Roberrson. 19g6.Parental care of nestlings by male red_wingedblackbirds. Behavioral Ecology and SociobiolJgy19:105*114.

Nelms, C. O., D. L. Otis, G. M. Linz, and W. J. Bleier.1999. Cluster sampling to estimate breedingblackbird populations in North Dakora. WildlifeSociety Bul lerin 27 :93 1_937 .

SAS Institute, Inc. 1999. SAS/STAT user,s guide, version8. SAS Institute,Inc., Cary, North Caiolina, USA.

Searcy, W. A:, and K. yasukawa. lgg5. polygyny andsexual selection in red_winged blackbirds.Princeton University press, princeion, New Jersey,USA.

Stoehr, A. M. 1999. Are significance thresholdsappropriate for the study of animal behaviour.Animal Behaviour 57 :F22_F25

The Prairie Naturalist' 43Q/D: December 2011

Teather, K.L. 1992. An experimental study of competitionfor food between male and female nestlings of thered-winged blackbird. Behavioral Ecology and

Sociobiology 3 1 :81-87.Webster, M. S. 1991. Male parental care and polygyny in

birds. American Naturalist I37 :27 4-280.Westneat, D. F., and P. W. Sherman. 1993. Parentage and

the evolution of parental behavior. BehavioralEcology 4:66J7.

Whittingham, L. A. 1989. An experimental study ofpaternal behavior in red-winged blackbirds.Behavioral Ecology and Sociobiology 25 :7 3-80 .

99

Whittingham,L.A., and R. J. Robertson. 1994. Foodavailability, parental care, and male mating success

in red-winged blackbirds (Agelaius phoeniceus).Journal of Animal Ecology 63:139-150.

Yasukawa, K., and W. A. Searcy. 1995. Red-winged, blackbird (Agelaius phoeniceus). Number 184 rn

,A. Poole and F. Gill, editors. The birds of NorthAmerica. The Aqademy of Natural Sciences and

the American Ornithologists' Union. Philadelphia,Pennsylvania, USA and Washington, D.C., USA.

Submitted 4 April 2011. Accepted 28 August 2011.Associate Editor was C. P. Lehman.